- Full text PDF available (5)
This paper describes a method by which a computer can autonomously acquire training data for learning to recognize a user’s face. The computer, in this method, actively seeks out opportunities to acquire informative face examples. Using the principles of co-training, it combines a face detector trained on a single input image with tracking to extract… (More)
Relational databases are the current standard for storing and retrieving data in the pharmaceutical and biotech industries. However, retrieving data from a relational database requires specialized knowledge of the database schema and of the SQL query language. At Anadys, we have developed an easy-to-use system for searching and reporting data in a… (More)
Goal: fast, accurate googling on video files (movies). Uses document-retrieval methods. Arbitrary visual-content queries execute very quickly. A video frame is analogous to a document. A visual interest point is analogous to a word. Video data is analyzed and indexed once. Each movie is treated as a separate database.
This paper presents an extension to David Lowe's well-known object recognition algorithm based on his Scale and Feature Invariant Transform (SIFT). One of the benefits of Lowe's SIFT-based method is that it can recognize objects from only three keypoints. While this capability can be useful in circumstances where the cost of a false negative is high, it's… (More)
While established pharmaceutical companies have chemical information systems in place to manage their compounds and the associated data, new startup companies need to implement these systems from scratch. Decisions made early in the design phase usually have long lasting effects on the expandability, maintenance effort, and costs associated with the… (More)
A device to facilitate accurate laboratory planning of maxillary osteotomies is described. The modification of a standard Dentatus articulator can easily be fashioned by a maxillofacial technician.
Find the best node-to-node matching between two attributed graphs. a b c d e f g q h i j k l m n o p May have missing or extra nodes. Graphs may be rotated relative to one another. Both graphs are fully connected.
The application context is visual object recognition in a social robot.